high-context local setup: anyone testing this?

[thetashed24]

Maybe I'm missing something obvious.

I'm trying to decide whether to keep the current stack or change course. Main question is how people here are handling embeddings.

If you've tested something similar, what did you keep and what did you stop bothering with?

I care more about stability than squeezing the last possible tokens/sec out of it.

[mangovault67]

A couple things I'd check first. If you're on TensorRT-LLM, I'd confirm the version first and then reduce variables until the behavior is obvious. Once you know the plain path works, then add the faster path back in piece by piece.

[batchshed]

That matches what I was seeing. I don't think your conclusion is far off.

[modbench]

Interesting, I hadn't tried that. I ended up in about the same place using TensorRT-LLM. I got better results once I stopped chasing one metric and just looked at whether the workflow felt usable.

[torchsmith25]

I'd approach it in a boring order. If you're on TensorRT-LLM, I'd confirm the version first and then reduce variables until the behavior is obvious.

[macronest]

A couple things I'd check first. On 4090, the easiest win is often to back off the aggressive settings and see whether the problem survives. Once you know the plain path works, then add the faster path back in piece by piece.

[tinycontext]

A couple things I'd check first. On 4090, the easiest win is often to back off the aggressive settings and see whether the problem survives. Once you

Tried to reproduce it just now. On 5090 with vLLM, I saw basically the same behavior. I expected the model choice to matter more than it did.

[thetarunner]

[quote]A couple things I'd check first. On 4090, the easiest win is often to back off the aggressive settings and see whether the problem survives. On

Thanks, that helps. I don't think your conclusion is far off. I got better results once I stopped chasing one metric and just looked at whether the workflow felt usable.

I'm happy to test one more variant if you want a second data point.

[amberplane]

Thanks, that helps. That matches what I saw on 5090.

[cachebench]

that fixed it for me too.